-
Notifications
You must be signed in to change notification settings - Fork 1
/
run_language_modeling.py
362 lines (322 loc) · 16.5 KB
/
run_language_modeling.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
#!/usr/bin/env python
# coding=utf-8
# Copyright 2020 The HuggingFace Team All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""
Explanation Generation by prefix-tuning
"""
import logging
import sys
import os
import pickle
from torch.optim import Adam, AdamW
from torch.optim.lr_scheduler import ConstantLR
import transformers
from transformers import (
AutoConfig,
HfArgumentParser,
TrainingArguments,
Trainer,
EvalPrediction,
BartTokenizerFast,
GPT2TokenizerFast,
set_seed
)
from transformers.trainer_utils import get_last_checkpoint
# import datasets
from arguments import ModelArguments, DataTrainingArguments
from data_utils import load_table2text_dataset, DataCollatorForTable2Text
from model.modeling_bart import BartForConditionalGeneration
from model.modeling_gpt2 import GPT2LMHeadModel
from model.bart_lm_prefix_model import PrefixTuning_BartforLM
from model.gpt2_lm_prefix_model import PrefixTuning_GPT2ForLM
from trainer import EvalPrediction, EvaluateFriendlySeq2SeqTrainer
from metrics import compute_bleu, compute_bleu_metric, compute_rouge_metric
logger = logging.getLogger(__name__)
def main():
os.environ["WANDB_DISABLED"] = 'true'
# See all possible arguments in src/transformers/training_args.py
# or by passing the --help flag to this script.
# We now keep distinct sets of args, for a cleaner separation of concerns.
# All these arguments can be found in arugments.py
parser = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments))
if len(sys.argv) == 2 and sys.argv[1].endswith(".json"):
# If we pass only one argument to the script and it's the path to a json file,
# let's parse it to get our arguments.
model_args, data_args, training_args = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1]))
else:
model_args, data_args, training_args = parser.parse_args_into_dataclasses()
logging.basicConfig(
format="%(asctime)s - %(levelname)s - %(name)s - %(message)s",
datefmt="%m/%d/%Y %H:%M:%S",
handlers=[logging.StreamHandler(sys.stdout)],
)
logger.info(model_args)
logger.info(data_args)
log_level = logging.INFO
logger.setLevel(log_level)
transformers.utils.logging.set_verbosity(log_level)
transformers.utils.logging.enable_default_handler()
transformers.utils.logging.enable_explicit_format()
# Log on each process the small summary:
logger.warning(
f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}"
+ f"distributed training: {bool(training_args.local_rank != -1)}, 16-bits training: {training_args.fp16}"
)
logger.info(f"Training/evaluation parameters {training_args}")
# Detecting last checkpoint.
last_checkpoint = None
if os.path.isdir(training_args.output_dir) and training_args.do_train and not training_args.overwrite_output_dir:
last_checkpoint = get_last_checkpoint(training_args.output_dir)
if last_checkpoint is None and len(os.listdir(training_args.output_dir)) > 0:
raise ValueError(
f"Output directory ({training_args.output_dir}) already exists and is not empty. "
"Use --overwrite_output_dir to overcome."
)
elif last_checkpoint is not None and training_args.resume_from_checkpoint is None:
logger.info(
f"Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change "
"the `--output_dir` or add `--overwrite_output_dir` to train from scratch."
)
# Set seed before initializing model.
set_seed(training_args.seed)
if data_args.dataset_file is not None:
with open(data_args.dataset_file, 'rb') as fin:
dataset = pickle.load(fin)
if training_args.do_train:
train_dataset = load_table2text_dataset(
dataset.input_ids_train, dataset.attention_mask_train, dataset.labels_train, dataset.item_labels_train, dataset.user_labels_train, data_args.max_train_samples
)
if training_args.do_eval:
eval_dataset = load_table2text_dataset(
dataset.input_ids_eval, dataset.attention_mask_eval, dataset.labels_eval, dataset.item_labels_eval, dataset.user_labels_eval, data_args.max_eval_samples
)
if training_args.do_predict:
test_dataset = load_table2text_dataset(
dataset.input_ids_test, dataset.attention_mask_test, dataset.labels_test, dataset.item_labels_test, dataset.user_labels_test, data_args.max_predict_samples
)
else:
raise ValueError("Please specify the dataset file")
config = AutoConfig.from_pretrained(
model_args.config_name if model_args.config_name else model_args.model_name_or_path,
cache_dir=model_args.cache_dir,
revision=model_args.model_revision,
use_auth_token=True if model_args.use_auth_token else None,
)
if model_args.model_type == 'gpt2':
config.pad_token_id = 50257
elif model_args.model_type == "bart":
config.pad_token_id = 1
if model_args.tuning_mode == "prefixtune":
if model_args.prefix_seq_len == 0:
raise ValueError("Please specify prefix_seq_len")
if model_args.mid_dim == 0:
raise ValueError("Please specify mid_dim (512 in general)")
if model_args.with_interaction is None:
raise ValueError("Please specify whether the interaction layer should be added")
config.preseqlen = model_args.prefix_seq_len
config.mid_dim = model_args.mid_dim
config.with_interaction = model_args.with_interaction
config.tuning_mode = "prefixtune" # useless? probably drop in future
config.add_item_prefix = model_args.add_item_prefix
elif model_args.tuning_mode == "finetune":
config.tuning_mode = "finetune" # same as above
else:
raise ValueError(f"Unrecoginized tuning_mode {model_args.tuning_mode}")
if model_args.num_users != 0:
config.num_users = model_args.num_users
else:
config.num_users = len(dataset.user_id_list)
logger.warning("You are giving num_users based on the dataset, it's better to check and specify by yourself.")
if model_args.num_items != 0:
config.num_items = model_args.num_items
else:
config.num_items = len(dataset.item_id_list)
logger.warning("You are giving num_items based on the dataset, it's better to check and specify by yourself.")
if model_args.model_type == "gpt2":
if model_args.tuning_mode == "finetune":
model = GPT2LMHeadModel.from_pretrained(
model_args.model_name_or_path,
from_tf=bool(".ckpt" in model_args.model_name_or_path),
config=config,
cache_dir=model_args.cache_dir,
revision=model_args.model_revision,
use_auth_token=True if model_args.use_auth_token else None,
)
# We add one additional [PAD] token during the tokenization. Need to resize the embedding.
model.resize_token_embeddings(50258)
logger.info("Initialize Fine Tuning GPT2 for LM successfully")
elif model_args.tuning_mode == "prefixtune":
if model_args.model_name_or_path == "gpt2":
pretrained_model = GPT2LMHeadModel.from_pretrained(
model_args.model_name_or_path,
from_tf=bool(".ckpt" in model_args.model_name_or_path),
config=config,
cache_dir=model_args.cache_dir,
revision=model_args.model_revision,
use_auth_token=True if model_args.use_auth_token else None,
)
# We add one additional [PAD] token during the tokenization. Need to resize the embedding.
pretrained_model.resize_token_embeddings(50258)
# Freeze Bart Parameter
logger.info("Freeze GPT2 parameters")
for n, param in pretrained_model.named_parameters():
# Only freeze the lm part, not the head.
if "transformer" in n:
param.requires_grad = False
model = PrefixTuning_GPT2ForLM(config, gpt2_model=pretrained_model)
logger.info("Initialize Prefix Tuning GPT2 for LM successfully")
else:
model = GPT2LMHeadModel.from_pretrained(
model_args.model_name_or_path,
from_tf=bool(".ckpt" in model_args.model_name_or_path),
config=config,
cache_dir=model_args.cache_dir,
revision=model_args.model_revision,
use_auth_token=True if model_args.use_auth_token else None,
)
logger.info("Load Prefix Tuning GPT2 for LM successfully")
elif model_args.model_type == "bart":
if model_args.tuning_mode == "finetune":
model = BartForConditionalGeneration.from_pretrained(
model_args.model_name_or_path,
from_tf=bool(".ckpt" in model_args.model_name_or_path),
config=config,
cache_dir=model_args.cache_dir,
revision=model_args.model_revision,
use_auth_token=True if model_args.use_auth_token else None,
)
logger.info("Initialize Fine Tuning Bart for LM successfully")
elif model_args.tuning_mode == "prefixtune":
if model_args.model_name_or_path == "facebook/bart-base":
pretrained_model = BartForConditionalGeneration.from_pretrained(
model_args.model_name_or_path,
from_tf=bool(".ckpt" in model_args.model_name_or_path),
config=config,
cache_dir=model_args.cache_dir,
revision=model_args.model_revision,
use_auth_token=True if model_args.use_auth_token else None,
)
# Freeze Bart Parameter
logger.info("Freeze Bart parameters")
for n, param in pretrained_model.named_parameters():
# Only freeze the lm part, not the head.
if "model" in n:
param.requires_grad = False
model = PrefixTuning_BartforLM(config, bart_model=pretrained_model)
logger.info("Initialize Prefix Tuning Bart for LM successfully")
else:
model = PrefixTuning_BartforLM.from_pretrained(
model_args.model_name_or_path,
from_tf=bool(".ckpt" in model_args.model_name_or_path),
config=config,
cache_dir=model_args.cache_dir,
revision=model_args.model_revision,
use_auth_token=True if model_args.use_auth_token else None,
)
logger.info("Load Prefix Tuning Bart for LM successfully")
else:
raise ValueError("We only allow gpt2 and bart as our base model.")
# lm_blue_metric = datasets.load_metric("bleu")
# lm_rouge_metric = datasets.load_metric("rouge")
def compute_metrics(eval_predictions: EvalPrediction, section: str):
# Compute BLEU score
# bleu_4 = lm_blue_metric.compute(predictions = eval_predictions.predictions, references=eval_predictions.items)
# bleu_1 = lm_blue_metric.compute(predictions = eval_predictions.predictions, references=eval_predictions.items, max_order=1)
bleu_4 = compute_bleu_metric(predictions = eval_predictions.predictions, references=eval_predictions.items)
bleu_1 = compute_bleu_metric(predictions = eval_predictions.predictions, references=eval_predictions.items, max_order=1)
output = {"bleu_1": bleu_1["bleu"], "bleu_4": bleu_4["bleu"]}
# Compute Rouge score
predictions_str = [' '.join(pred).strip() for pred in eval_predictions.predictions]
references_str = [' '.join(ref[0]).strip() for ref in eval_predictions.items]
# rouge = lm_rouge_metric.compute(predictions=predictions_str, references=references_str)
rouge = compute_rouge_metric(predictions=predictions_str, references=references_str)
rouge_1 = {
"r1_p": rouge["rouge1"].mid.precision,
"r1_r": rouge["rouge1"].mid.recall,
"r1_f": rouge["rouge1"].mid.fmeasure
}
rouge_2 = {
"r2_p": rouge["rouge2"].mid.precision,
"r2_r": rouge["rouge2"].mid.recall,
"r2_f": rouge["rouge2"].mid.fmeasure
}
output = {**output, **rouge_1, **rouge_2}
return output
data_collator = DataCollatorForTable2Text(
tuning_mode = model_args.tuning_mode,
add_item_prefix = model_args.add_item_prefix,
prefix_only = model_args.prefix_only
)
params = [p for p in model.parameters() if p.requires_grad]
if model_args.tuning_mode == "prefixtune":
name_params = [n for n, p in model.named_parameters() if p.requires_grad]
logger.info(f"All params that will be optimized are {name_params}")
optimizer = AdamW(params=params, lr=training_args.learning_rate)
# scheduler = StepLR(optimizer, step_size=0, gamma=0.0)
scheduler = ConstantLR(optimizer)
if "gpt2" == model_args.model_type:
tokenizer = GPT2TokenizerFast.from_pretrained("gpt2")
tokenizer.add_special_tokens({'pad_token': '[PAD]'})
elif 'bart' == model_args.model_type:
tokenizer = BartTokenizerFast.from_pretrained("facebook/bart-base")
trainer = EvaluateFriendlySeq2SeqTrainer(
model=model,
args=training_args,
data_collator=data_collator,
train_dataset=train_dataset if training_args.do_train else None,
eval_dataset=eval_dataset if training_args.do_eval else None,
eval_examples=dataset.review_text_eval if data_args.max_eval_samples is None else dataset.review_text_eval[:data_args.max_eval_samples],
optimizers=(optimizer, scheduler),
compute_metrics=compute_metrics,
tokenizer=tokenizer,
)
if training_args.do_train:
checkpoint = None
if training_args.resume_from_checkpoint is not None:
checkpoint = training_args.resume_from_checkpoint
elif last_checkpoint is not None:
checkpoint = last_checkpoint
train_result = trainer.train(resume_from_checkpoint=checkpoint)
trainer.save_model() # Saves the tokenizer too for easy upload
metrics = train_result.metrics
max_train_samples = (
data_args.max_train_samples if data_args.max_train_samples is not None else len(train_dataset)
)
metrics["train_samples"] = min(max_train_samples, len(train_dataset))
trainer.log_metrics("train", metrics)
trainer.save_metrics("train", metrics)
trainer.save_state()
# Evaluation
if training_args.do_eval:
logger.info("*** Evaluate ***")
metrics = trainer.evaluate()
max_eval_samples = data_args.max_eval_samples if data_args.max_eval_samples is not None else len(eval_dataset)
metrics["eval_samples"] = min(max_eval_samples, len(eval_dataset))
trainer.log_metrics("eval", metrics)
trainer.save_metrics("eval", metrics)
if training_args.do_predict:
logger.info("*** Predict ***")
predict_results = trainer.predict(
test_dataset=test_dataset if test_dataset else eval_dataset,
test_examples=dataset.review_text_test if data_args.max_predict_samples is None else dataset.review_text_test[:data_args.max_predict_samples],
metric_key_prefix="predict"
)
metrics = predict_results.metrics
max_predict_samples = len(test_dataset)
metrics["predict_samples"] = min(max_predict_samples, len(test_dataset))
trainer.log_metrics("predict", metrics)
trainer.save_metrics("predict", metrics)
if __name__ == "__main__":
main()